pipeline-and-forecast
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied sales data from external JSON files or command-line arguments, creating a potential surface for indirect prompt injection if the output is consumed by an agent without further validation.
- Ingestion points: The
resources/forecast.jsscript reads pipeline data (deals, stages, amounts) from local files viafs.readFileSyncor from inline JSON arguments. - Boundary markers: The skill does not implement specific delimiters or 'ignore' instructions for the data it processes.
- Capability inventory: The script is limited to local file reading and console output; it contains no network or shell execution capabilities.
- Sanitization: The script performs basic data validation, ensuring amounts are numeric and stages are mapped to known probabilities, and provides descriptive error messages for invalid input.
Audit Metadata